If model run to un-implemented layer using GPU or DSP, will switch to CPU runtime(Of course, this layer CPU implemented)?
If un-implemented layer using GPU or DSP, will switch to CPU runtime?
Posted: Sun, 2019-03-24 08:47
If model run to un-implemented layer using GPU or DSP, will switch to CPU runtime(Of course, this layer CPU implemented)?
Opinions expressed in the content posted here are the personal opinions of the original authors, and do not necessarily reflect those of Qualcomm Incorporated or its subsidiaries (“Qualcomm”). The content is provided for informational purposes only and is not meant to be an endorsement or representation by Qualcomm or any other party. This site may also provide links or references to non-Qualcomm sites and resources. Qualcomm makes no representations, warranties, or other commitments whatsoever about any non-Qualcomm sites or third-party resources that may be referenced, accessible from, or linked to this site.
Hi ysh329,
Yes, SNPE has CPU fallback mode which is exactly same feature as what you mentioned. Refer to below API document.
https://developer.qualcomm.com/docs/snpe/group__c__plus__plus__apis.html#ab4fe2d706b18da34ea98f92f1b7fb3cf
Thanks,
jihoonk